The broad objectives of this proposal are two-fold. One is to determine more precisely the brain origins of different types of visual evoked potentials (VEP) in order to further the theoretical understanding of visual brain organization. The methods involve simultaneous recording from multiple scalp electrodes of visual evoked potentials in response to two types of visual pattern stimulation (onset-offset and reversal), as well as to unpatterned flash. Using lateral hemifield stimulation and isopotential scalp contour mapping at all sequential latencies int he VEPs, global field power and spatial segmentation analyses will be performed. For stable periods, equivalent dipole sources will be computed. Latencies of particular interest will be that of an 80 msec component (isolated by bipolar recording) and the P100 (positive peak at 100 msec) for pattern reversal stimulation. Latencies of particular interest in pattern onset stimulation will be those of C1 (positive peak at about 80 msec) and C11 (about 100 msec). The cruciform model (calcarine and medial fissures) of primary visual cortex (area 17) will be investigated. Foveal and peripheral stimulation will be employed. The number of phase reversals obtained when recording VEPs at a fixed position from a small stimulus area at various angles from fixation will also be studied. The objective is the three dimensional localization within the brain of the sites of origin of the various VEP components (primary or secondary visual cortex or other brain areas). The second objective is the better localization of loci of visual brain lesions, which is unreliably done by present pattern reversal (P100) and blank flash VEP techniques. As with normals, at all latencies, isopotential contour maps, global field power, and spatial segmentation analysis will be calculated. For stable periods, equivalent dipole sources will be computed. Latencies of particular interest are Cl of pattern-onset (little used for this purpose) and the 80 msec component from pattern reversal. Patients to be studied will include those with visual brain lesions from vascular accident, trauma, and surgical intervention such as hemispherectomies. With the methods from the first objective, better localization of the etiologic brain lesions can be provided.